compositions and phase diagrams for aqueous systems …...compositions and phase diagrams for...

Compositions and Phase Diagrams for Aqueous Systems Based on Proteins and Polysaccharides Vladimir Tolstoguzov Nest16 Research Centre, CH-1000 Lausanne 26, Switzerland

~

Limited thermodynamic compatibility of proteins with other proteins and proteins with polysaccharides is a fundamental phenomenon that has been demonstrated in more than 200 biopolymer pairs. These systems can undergo a liquid-liquid phase separation resulting in the different macromolecular components primarily concentrated in the different phases. This occurs under conditions (pH values and ionic strengths) inhibiting attraction between nonidentical biopolymers, i.e., the formation of interbiopolymer complexes. Generally, phase separation takes place when the total concentration of the macromolecular components exceeds a certain critical value. The excluded volume of the macromolecules determines both their thermodynamic activity and phase separation threshold. Phase diagrams of biopolymer mixtures and physicochemical features of biphasic systems are considered here. Attention is centered on the limited compatibility of the main classes of proteins and various polysaccharides and on the effects of variables such as pH, ionic strength, temperature and shear forces on the phase state, equilibrum and structure of these two-phase liquid systems. The general nature of the phenomenon of thermodynamic incompatibility of biopolymers accounts for its importance in structure formation in cytoplasm.

KEY WORDS: Biopolymer incompatibility, Excluded volume, Phase behavior of biopolymer solutions, Phase diagram, Protein-salt-water systems, Protein-protein-water systems, Protein-polysaccharide-water systems, Membraneless osmosis, Biphasic system features.

1. Introduction: Limited Compatibility of Macromolecular Compounds

The history of the experimental study of polymer incompatibility, like the history of the whole polymer science, started with natural macromolecules.

International Review of Cytology, VoI. 192 3 Copyright 0 2ooo by Academic Press. All rights of reproduction in any form reserved. 0074-7696100 $30.00

4 VL4DlMlR TOLSTOGUZOV

The first observation of a limited thermodynamic compatibility (or a limited cosolubility) of biopolymers (by which is meant proteins or polysaccharides) was published more than 100 years ago. Professor W. Beijerinck (1896, 1910) discovered the impossibility of mixing biopolymers in the common solvent, water. The two solutions of gelatin and starch formed a water-in- water emulsion instead of a homogeneous mixture. This emulsion settled into two liquid layers. One of the layers was the transparent gelatin solution containing a small quantity of starch while the other layer contained starch and a small amount of gelatin. Ostwald and Hertel (1929) then found a difference in phase behavior between cereal and potato starches in mixtures with gelatin. They showed that, after phase separation, the liquid phase rich in starch turns into a powder-like precipitate. Doi and Nikuni (1962) discovered the crystallization of amylopectin in mixtures with gelatin. They hypothesized that this phenomenon could be a model for starch deposition in the plant cell. Using the cloud point method, Doi (1965) constructed first phase diagrams for the gelatin-amylopectin-water system using both temperature-gelatin content and gelatin-amylopectin contents as coordinates. Bungenberg de Jong (1936) showed that on mixing aqueous solutions of gelatin and gum arabic two different types of phase separation could occur. He determined the first phase diagrams for both phase separation types and gave them the names “simple” and “complex coacervation.” Simple coacervation results in concentration of the biopolymers in different phases, i.e., it corresponds to biopolymer incompatibility. It takes place at a sufficiently high ionic strength and pH values above the isoelectric point (IEP) of gelatin where the biopolymer macro-ions have like net charges. Complex coacervation is the formation of interbiopolymer complexes, i.e., phase separation results in concentration of biopolymers into a single liquid, highly hydrated, concentrated phase with a solution of either an excess of one of the macro-ions or a nearly pure solvent as the other phase. For instance, at pHs below gelatin’s IEP and at a low ionic strength electrostatic complexes of the positively charged macro-ions of gelatin and the negatively charged macro-ions of gum arabic are formed. Dobry and Boyer-Kawenoki (2948) showed that phase separation occurs in mixed solutions of many water-soluble polymers such as gelatin, serum albumin, gum arabic, glyco- gen, polyvinyl alcohol, polyvinylpyrrolidone, methylcellulose, polyacrylic acid, etc. Later, phase behavior of mixtures of gelatin with polysaccharides were studied by Grinberg et al. (1970, 1971), Grinberg and Tolstoguzov (1972, 1997), Tolstoguzov et al. (1974a,b, 1985), Tolstoguzov (1990, 1991, 1992), Kasapis et al. (1993), and Clewlow el al. (1995).

During the last 50 years the major efforts in the field of polymer compatibility were stimulated by practical problems related to mixtures of the following macromolecular compounds.

PHASE DIAGRAMS FOR AQUEOUS SYSTEMS 5

(i) Polymer blends are of great applied importance for controlling mechanical, optical, and other physical properties of materials such as plastics, rubbers, films, fibers, glues, etc. Normally, chemically or structurally dissimilar polymers have limited thermodynamic compatibility or limited miscibil- ity, i.e., for thermodynamic reasons (low mixing entropy of macromolecules), these polymers are immiscible with each other on a molecular level. The Flory-Huggins (Flory, 1953; Tanford, 1961) theory of polymer solutions, developed for flexible linear and neutral polymers, predicts their phase behavior. Polymers form single-phase mixed solutions only when their mixing process is exothermic. This type of polymer pair is rarely encountered (Krause, 1978; Kwei and Wang, 1978). It was shown, theoreti- cally and experimentally, that the transition from polymer mixtures to polyelectrolyte mixtures increases compatibility. This enhancement of the cosolubility of polyelectrolytes reflects an increase in mixing entropy due to contributions of low molecular weight counterions under electrically neutral conditions (Khokhlov and Nyrkova, 1992).

(ii) Mixtures of polysaccharides and synthetic water-soluble polymers are of practical significance in the pharmaceutical industry and biotechnology . Albertsson (1958, 1972) was a pioneer in the systematic study of phase behavior of aqueous mixed solutions of polysaccharides, modified polysaccharides, and synthetic polymers. Applications of these systems for fractionation and purification of biomaterials, such as cells, cell organelles, and biopolymers, have been developed and discussed by Albertsson (1995), Walter et aZ. (1985,1991), and Walter and Brooks (1995). In the pioneering works of Ogston (1937) and Laurent and Ogston (1963), an original approach for estimating thermodynamic interactions and compatibility of water-soluble polymers was developed. This approach was based on the excess of osmotic pressure in mixed solutions of incompatible polymers and used for the study of biphasic systems, as proposed by Albertsson. The authors assumed that phase separation of biopolymers occurs in biological systems (Edmond and Ogston, 1968).

(iii) Protein-salt-water systems are of interest in the medical (ophthalmol- ogy) and food technologies although they may be less relevant to the possibility of phase separation in native cytoplasm. Pioneering investigations on protein-salt-water systems were camed out on lactoglobulin and seed storage globulins (including arachin and soybean globulins) by Tombs at Unilever Research in the 1960s-1970s (Tombs, 1970a,b, 1972,1975,1985; Tombs et al., 1974) and then on lens crystallins by Benedek et al. at M.I.T. (Ashene et al., 1996; Lomakin et aZ., 1996; Liu et aZ., 1995; Broide et al., 1991). Phase separation in salt solutions of seed storage globulins was systematically studied in several other laboratories (Ismond et al., 1988, 1990, Popello et aL, 1990,1991,1992; Suchkov et aZ., 1990,1997; Tolstoguzov,

6 VLADlMlR TOLSTOGUZOV

1988a, 1991, 1992). The formation of highly concentrated noncrystalline phases is of significance for both lens crystallins and seed storage globulins.

(iv) Biopolymer mixtures are of importance for controlling the composition-structure-property relationship in formulated foods (Grinberg and Tol- stoguzov, 1972, 1997; Kasapis et a!., 1993; Moms, 1990; Polyakov et aL, 1997; Tolstoguzov et al., 1985; Tolstoguzov, 1978,1986,1988b, 1991,1993b7 1996a,b, 1997a,b). Proteins and polysaccharides are the main structure forming food materials. Each food contains biopolymer mixtures. Liquid aqueous biphasic systems, i.e., water-in-water emulsions, are typical of foods. The term a “water-in-water emulsion” was invented (Tolstoguzov, 1986,1988a,b) to distinguish them from oil-in-water and water-in-oil emulsions (Tolstoguzov et af., 1974b).

Systematic investigations in the areas of protein-polysaccharide mixtures and protein mixtures started in the laboratory of novel food forms of the USSR Academy of Sciences about 30 and 20 years ago, respectively. These works were reviewed by Grinberg and Tolstoguzov (1997), Ledward (1993), Polyakov et ai. (1997), Samant et af. (19931, Tolstoguzov et af. (1985), and Tolstoguzov (1978, 1986, 1988c, 1991, 1997aj. Until that time information on the thermodynamic compatibility of biopolymers had been confined to gelatin-polysaccharide mixtures. Gelatin, however, behaves like classical flexible linear chain polymers, i.e., unlike both compact globular proteins and rigid polysaccharides. The next investigations were into the polyelectrolyte nature of biopolymers and the ability of neutral polysaccharides to bind low molecular weight ions in aqueous solutions. However, biopolymer mixtures remained experimentally unstudied for a long time. For instance, it was not clear at all whether aqueous solutions of globular proteins and their mixtures with polysaccharides could undergo phase separation. By contrast, the formation of interbiopolymer complexes, e.g., by oppositely charged globular proteins with each other and with polysaccharides, i.e., complex coacervation, was well known and attracted more attention (Led- ward, 1993; Poole et al., 1984, Tolstoguzov, 1978, 1986, 1990, 1994b).

Phase state analysis of more than 200 biopolymer pairs showed the general nature of their thermodynamic incompatibility in a common aqueous solvent. Experimental studies have shown that incompatibility occurs in the following biopolymer mixtures: (i) proteins and polysaccharides; (ii) proteins belonging to different classes according to Osborne’s classification (i.e., albumins, globulins, glutelines, and prolamines j; (iii) native and denatured proteins; (iv) aggregated and nonaggregated forms of the same protein; and (v) structurally dissimilar polysaccharides (Antonov et al., 1975, 1979, 1980, 1982, 1987; Grinberg et aL, 1970, 1971; Grinberg and Tolstoguzov, 1972, 1997; Kalichevsky et al., 1986, Kalichevsky and Ring, 1987; Polyakov et al., l979,1980,1985a,b, 1986a,b, 1997; Tolstoguzov, 1978, 1992, 1997a). Incompatibility also probably occurs between dissolved and


adsorbed (at the oiYwater interface) forms of the same protein. Denatur- ation and partial hydrolysis of proteins oppositely influence their incompatibility with other biopolymers (Tolstoguzov, 1991).

Since both biological and food systems generally contain the same propor- tions of macromolecular components, information about phase behavior of foods is of great interest for understanding that in cytoplasm. However, phase separation probably occurs more often in food than in biological systems. The reason seems to be that the biological systems are mainly based on specific interactions (mainly attraction) of biopolymers, while food is based on nonspecific interactions (both attraction and repulsion) of macromolecular compounds.

The aim of this chapter is to consider phase behavior of biopolymer mixtures, since this information is of importance in the new concept of phase separation in cytoplasm proposed by Walter and Brooks (1995).

II. Phase Diagrams

The phase behavior of a mixed biopolymer solution is quantitatively charac- terized by a phase diagram describing the boundary conditions of phase separation and the partitioning of the components (i.e., the water and the biopolymers) between the phases. Phase diagrams are used for graphically presenting effects of temperature, pH, salt concentration, and other variables on phase behavior of biopolymers. Two types of liquid-liquid phase separation, namely, in salt solutions of proteins and in mixed biopolymer solutions. will be considered.

A. Phase Behavior of ProteinSalt-Water Systems

Figure 1 gives the typical phase diagrams of the 11s broad bean globulin (legumin) in an aqueous salt solution. These phase diagrams plot: (a) salt concentration versus protein concentration; (b and c) temperature versus protein concentration. The latter resemble the behavior observed for lens crystallin described in the chapter by Clark and Clark in this volume. They show a reversible transition of salt legumin solutions from the single-phase to the two-phase state on diluting with water (a), cooling (b) and shearing (c) (Popello er al., 1990,1991,1992; Suchkov er al., 1990,1997; Tolstoguzov, 1988b, 1992, 1996b). The solid lines are the binodal (coexistence) curves and present compositions of the coexisting phases. The regions lying above and below the binodal curves represent biphasic and single-phase systems, respectively. The right and left branch of each binodal correspond to the


0.51 . 0.45

D

I 0 ' b

0 Ccrm 40 Protein, O/wt

0 m 40 Protein, %wt.

a

b

C

FIG. 1 Phase diagrams of the 11s globulin (legumin) of broad bean-sodium chloride-water system: (a) pH 4.8, T = 8°C; (b) pH 4.8, 0.6 M NaCI; (c) pH 4.8, 0.6 M NaCI; the binodals (I) and (n) are determined by turbidimetric and rheological techniques, respectively.

composition of the concentrated bottom phase (called mesophase) and the less concentrated top phase. The thin horizontal lines are the tie-lines. Each tie-line connects three points representing the compositions of the system as a whole and of both equilibrium phases. The relative amount of the coexisting phases is estimated by the inverse lever rule. The phase volume ratio corresponds to the ratio of the tie-line segments: CBBD. Figures la and b show that on cooling or dilution of the initial protein solution A with water or its dialysis against water to the point B results in phase


separation into two liquid phases, phase D and phase C. The protein concentration in the bottom phase can exceed 40%, while in the top phase it may be below 0.4%.

Since the phases strongly differ in protein concentration (and in refraction index), phase separation is accompanied by an increase in turbidity. There- fore, the cloud point technique can be used to determine the binodal points. Differences in concentration and density between the phases result in easy settling and formation of two liquid layers. Quantitation of the protein content in these layers gives the binodal and the tie-line positions. The phase separated turbid system C becomes transparent again after addition of a sufficient amount of salt (Fig. la) or by a sufficient increase in temperature (Fig. lb) or in shear forces (Fig. lc). The critical point shows either the maximum salt concentration (a) or the maximum temperature (b and c) at which phase separation occurs. At the critical point both coexisting phases have the same composition and volume (or mass, since their concentrations are low and densities are close to that of water). The dashed line coming vertically down from the critical points is the rectilinear diameter. It passes through the mid-tie-lines and represents the composition of systems demixing into phases of the same volume.

Systems containing native individual 11s and 7s seed globulins and their mixtures have upper critical points with the same critical protein concentration of 18%. The binodal curves are symmetrical about the rectilinear diameter. It has been assumed (Tolstoguzov, 1988c, 1991,1992) that phase separation occurs between associated and nonassociated forms of the same protein. This means that associated (micelle-like) and nonassociated forms of the same proteins cannot recognize each other as being the same. In other words, the difference in concentration between the protein-enriched and the protein-impoverished phases (D and C), which are in osmotic equilibrium, reflects a relatively high degree of protein association. It has been also suggested that the main factors determining the association of protein molecules and phase separation are the high excluded volumes of oligomeric seed storage proteins and the dipole-dipole interaction between their molecules. The attraction of large sized micelle-like associated protein particles suspended in aqueous solution of incompatible biopolymer probably arises from depletion flocculation (Tolstoguzov, l991,1994a, 1997a). It should be noted that milk casein, that also occurs as large, like-charged protein associates, has a similar phase behavior in mixtures with polysaccharides. Phase separation of mixtures of 11s globulin molecules and their associates is likely similar to the phase separation of the denatured (aggregated) and native forms of the same protein (ovalbumin) (Tolstoguzov, 1988b, 1991). The crystallization of seed storage proteins in the highly concentrated phase (mesophase) is probably inhibited by the presence of isomeric protein forms. Formation of micelle-like associates of protein

10 VMDlMlR TOLSTOGUZOV

molecules, i.e., a process competing with protein crystallization, is of importance for formation of “protein bodies,” which can be rapidly mobilized (hydrolyzed) during seed germination.

6. Phase Behavior of Biopolymerl -Biopolymer2- Water Systems

Conventionally, phase diagrams of three-component (ternary) systems are presented in triangular coordinates. Figure 2a shows a typical equilateral triangular diagram for a ternary system containing two biopolymers, A and B, and the solvent, water, C. Each comer of the equilateral triangle represents a pure component and its designation is marked at this corner. On the side opposite to this comer, the concentration of this component is zero. Each side of the triangle corresponds to a two-component system. The region inside the triangle represents mixtures of all three components. The system composition (e.g., X) is read along the axes, i.e., as a, b, and c coordinates whose sum equals unity.

However, an excess of solvent, water (compared to biopolymers), that is typical for biological systems, makes the use of phase diagrams in rectangular coordinates more practicable. This also simplifies plotting the effects of different variables. Figure 2b is a typical phase diagram for biopolymer- 1-biopolymer-2-water systems in rectangular coordinates. Normally, the concentrations of biopolymers are plotted on the axis in weight percent; the rest is assumed to be water. Every point in a phase diagram corresponds to a system composition. The bold curve is a binodal. Biopolymers are fully miscible with each other in the concentration region under the binodal.

b charide,% ,-1. _--- a PolyseGl

fraction Mass A M a s s fraction A&

of c Of B &?pear

’ Diameter

1 Binodal

FIG. 2 Typical phase diagram for a ternary biopolymer 1 - biopolymer 2 - water system presented in the form of (a) triangular and (b) rectangular coordinates.


For instance, on mixing aqueous solutions of a polysaccharide solution A and a protein B, a single-phase stable mixture of composition C may be obtained. Mixture compositions lying above the binodal curve correspond to two-phase systems, where the phases are aqueous mixed solutions exhib- iting limited co-solubility of biopolymers. For instance, on mixing a polysaccharide solution Al and protein solutions B or B,, mixtures of composition C1 and C, can be obtained. These mixed solutions spontaneously break down into two liquid phases, phase D and phase E. The phase D is rich in protein and another, E, is rich in the polysaccharide. On centrifugation, the protein-rich phase usually forms the lower liquid layer while the polysaccharide-rich phase forms the upper liquid layer. The composition of the more concentrated and denser bottom phase is usually plotted as the ab- scissa while the less dense top phase is presented as the ordinate.

The thin line DE is a tie-line. It connects the points representing the compositions of the coexisting equilibrium phases. Biopolymer mixed solutions whose compositions correspond to the same tie-line will break down into two phases of the same composition. For instance, all biopolymer mixtures of compositions DE will separate into the same phases D and E. The point of the initial biopolymer mixture (Cl) divides the tie-line in the two segments whose length ratio reflects the phase volume (weight) ratio. According to the inverse lever rule, the phase volume ratio corresponds to the ratio of the tie-line segments: EC1/CID. The tie-lines can be nonparal- lel since an increase in concentration of biopolymers is usually accompanied by their self-association. The rectilinear diameter is the dashed line passing through the mid-tie-lines. It gives the system compositions splitting into phases of the same volume. In the vicinity of the rectilinear diameter, phase inversion can occur. For instance, this takes place when the system’s composition changes from point C1 to C2. Lower values of the critical point F coordinates indicate lower co-solubility of the biopolymers and greater incompatibility. Table I gives critical point coordinates for several systems. The phase separation threshold G is the minimal total concentration of biopolymers required for phase separation to occur. The position of phase separation threshold geometrically corresponds to the point of contact between the binodal curve and the straight line cutting segments of the same length on the concentration axis of the biopolymers. The two- dimensional phase diagram corresponding to a certain constant temperature, pH, and ionic strength becomes three-dimensional when an additional variable, such as pH, salt concentration, or temperature is included (Tolsto- guzov et af., 1985).

C. Determination of Phase Diagrams

Normally, a preliminary qualitative study on phase behavior of a biopolymer mixture is very helpful. Techniques such as light microscopy, centrifuga-

TABLE I Critical Point Coordinates for Biopolymar Mixtures

Biopolymer pair

Critical point coordinates Protein, %wt.: Pr;

Polysaccharide, %wt.: Ps Incompatibility conditions

Gelatin (mol.wt.170,000) + Pectin (330,000, DE = 67.2%)

Gelatin (mol.wt.170,000) + alginate (400,ooO; MG = 50%) Gelatin (mol.wt.170,000) + alginate (400,000, MG = 20%) Gelatin (mol.wt.l70,000) + methylcellulose (70,000) Gelatin (moLwt.170,000) + dextran (2,000,000) Legumin (360,000) -t dextran (270,000) Legumin (360,000) + k-Carrageenate-Na (disordered conformation) Legumin (360,000) + k-Carrageenate-Na (helical conformation) Glycinin (360,000) + Pectinate-Na (DE = 58%) Gelatin + Viciu fubu globulins Gelatin + 11s Viciufaba globulin Gelatin + 11s Viciu fubu globulin Ovalbumin + Viciu fubu globulins Ovalbumin + thermodenatured ovalbumin

Pr-1.0; Ps-2.6 Pr-0.95; Ps-2.4 Pr-0.25; Ps-1.65 Pr-0.30; Ps-1.25 Pr-0.5; Ps-1.2

Pr-3.1; Ps-1.3 Pr-2.10-2; P~-5.10-~ Pr-5.10-3; 1.W Pr-9; Ps-0.28 Pr, -2.5; Pr2-11.6 Pr1-2.1; Pr2-12 Prl-4.0; Pr2-16.0 Pr, 40; Pr2 >15 Pr, - 10.8: Pr, - 3.6

PI-2.6 Ps-5.2

pH 6.0; 40°C pH 8.0; 40°C pH 6.0; 40°C 0.2 M NaCl pH 6.0; 40°C; 0.2 M NaCl pH 6.0; 40°C 0.2 M NaCl pH 6.0, 40°C; 0.2 M NaCl pH 7.8; 25°C; 0.1 M NaCl pH 7.8; 0.1 M NaCl; 40°C pH 7.8; 0.1 M NaC1; 10°C PH 7.8; 0.3 M NaCl pH 7.0; 40°C pH 7.0; 40°C pH 6.6; 40"C, NaCl 0.5 M" pH 6.6; 20°C PH 6.7; 20°C


tion, turbidimetry, and viscometry can be used. Single-phase and biphasic systems can be identified by visual inspection (by turbidity and demixing into two layers) and by light microscopy. Spherical and nonspherical dispersed particles observed by an optical microscope correspond to liquid and solid phases, respectively. The cloud point method for construction of a phase diagram is especially useful when initial solutions of individual biopolymers have a low optical density. Normally, either cloud-point tem- peratures or cloud-point concentrations of an added biopolymer are determined. For qualitative determination of the binodal, a solution of one of the biopolymers is added to a solution of the second one until turbidity occurs. Then the solution of the second biopolymer is added until the mixture becomes clear. These operations are repeated using series of the individual biopolymer solutions of different concentrations to determine the binodal position (Grinberg and Tolstoguzov, 1972). A turbodimetric titrator was developed for determination of cloud points corresponding to a certain value of the turbidity change at fixed temperature, pH and salt concentration (Antonov et d., 1975).

Viscosimetric titration can be used to construct phase diagrams of concentrated nontransparent systems (Tolstoguzov et al., 1969). Phase separation is usually accompanied by a decrease in viscosity (Suchkov er aL, 1997), which can be used for finding the binodal points (e.g., in Fig. lc, binodal 11). Viscosimetry was also used to study the interactions of biopolymers in dilute mixed solutions (Varfolomeeva et al., 1980).

Construction of phase diagrams usually starts with the preparation of series of mixed solutions sufficiently differing in bulk biopolymer concentration. Some of them can be single-phase solutions, others biphasic systems. A true equilibrium between the phases is experimentally obtained by mixing or shaking the water-in-water emulsions under different time-temperature conditions. Separation of the phases by centrifuge provides information about the state of matter and the volume ratio of the system phases. Centrif- ugation conditions vary for different water-in-water emulsions and can change greatly by phase inversion. The closer a system composition is to the critical point, the smaller the difference in density between the phases and the more difficult their separation by centrifugation.

The amount of each biopolymer in each phase separated by centrifugation can be quantified by various techniques. Estimation of protein concentration by UV absorbance at 280 nm (depending on the tyrosine, tryptophan, phenylalanine, and cysteine content) is widely used because of its simplicity and sensitivity. The disadvantage of the method is a strong contribution at the same wavelength from nucleic acids. An accurate method is binding of the anionic dye Coomassie blue to protein (to arginyl and lysyl side groups of the protein, but not to free amino acids) with an adsorbance maximum at 590 nm. The Biuret reaction, where the protein peptide bonds react with


Cu2+ under alkaline conditions, and the Lowry method, where the Folin reagent is additionally included, are not so widely applied because of strong interference from sugars, nucleic acids, and buffers. Kjeldahl’s method is specially applicable for measuring the percentage of proteins. The concentration of the polysaccharide is usually determined at 490 nm by the phenol sulfuric method (Dubois et al., 1956). Alternative methods have been recom- mended for analysis of the phases, e.g., optical activity, far-UV spectroscopy (190-220 nm), Fourier transform infrared spectroscopy, sedimentation, electrophoresis, HPLC, and other chromatographic and radiometric (radio- labelled biopolymer) techniques (Albertsson, 1972; Grinberg and Tolstogu- zov, 1972; Polyakov et ab, 1980; Durrani et al., 1993; Medin and Janson, 1993).

The phase volume ratio method was developed (Polyakov et al., 1980; Antonov ef al., 1987) for determination of phase diagrams without chemical analysis. In this case, the volumes of the two separated phases and the phase volume ratio are determined for a large number of mixed solutions whose composition presented in the weight fraction of one of the biopolymers varies from zero to unity. Then to determine two points of the binodal curve the experimental dependence of the system composition-phase volume ratio is graphically extrapolated to zero and unity values of the phase volume ratio. The system composition with 0.5 phase volume ratio corresponds to the point of the rectilinear diameter.

In all cases, the experimental data are usually checked by the material balance of equilibrium phase separation. The tie-lines have to be straight lines, each connecting three points representing the compositions of the system (e.g., C) and of its phases (e.g., D and E). The phase volume ratio has to correspond to the inverse lever rule. The critical point can be determined by extrapolation of the rectilinear diameter. The phase separation threshold is graphically determined as the point of contact of the binodal and the line intercepting equal lengths on the coordinate axes.

Ill. Factors Affecting Phase Behavior of Biopolymer Mixtures

A. Effects of Biopolymer Composition and Environmental Medium (Salt and Sugar Concentration, pH, and Temperature)

Mixed biopolymer solutions can be quantitatively thermodynamically char- acterized by the interactions of nonidentical macromolecules with each other and with the solvent in terms of the second virial coefficient, experi-


mentally obtained by light scattering. The cross second virial coefficient, reflects pair interactions of nonidentical biopolymers. Its positive value corresponds to a net repulsive biopolymer interaction and incompatibility. This approach has been used to study the interactions responsible for the incompatibility of different biopolymer pairs and the effect of additives, e.g., sucrose (Semenova et al., 1990, 1991a,b; Tolstoguzov, 1991, 1992, 1994a,b; Tsapkina el aL, 1992). For instance, it was found that thermodynamic compatibility of many biopolymer pairs (e.g., legumin and dextran, legumin and ovalbumin, sodium caseinate and ovalbumin, sodium caseinate and sodium alginate) increases with sugar concentration. Incompatibility occurs under certain conditions (pH values, ionic strengths, and temperature) favoring association of identical macromolecules and inhibiting attraction between nonidentical macromolecules, i.e., formation of interbiopolymer complexes. Unfavorable interactions (repulsion) of chemically and structurally nonidentical macromolecules result in each macromolecule preferring to be surrounded by its own type. Compatibility usually arises from the formation of weak soluble complexes with an energy of attractive interbiopolymer interaction of about 2 kT or higher (Varfolomeeva et al., 1980).

Most biopolymers are polyelectrolytes. Therefore, the major factors affecting interactions of biopolymers with each other and with the aqueous solvent are pH, salt concentration, the conformation and counterions of the biopolymers, their net charge, charge density, hydrophobicity, and H- bond forming ability. The counterions of biopolymers, as polyelectrolytes, contribute to an increase in mixing entropy and favor compatibility of biopolymers compared to classical polymers.

Normally, self-association of proteins increases in salt solutions when the pH approaches the protein’s IEP. This corresponds to an enhancement of thermodynamic incompatibility between proteins and polysaccharides. When the salt concentration is below a certain critical value, the protein and the polysaccharide are completely compatible. The compatibility of the macromolecular components drops sharply when the salt concentration exceeds the critical point. The critical salt concentration required for phase separation is a function of the pH and salt composition. It increases in the following order for polysaccharides: carboxyl- containing<neutral<sulfated.

Generally, protein-neutral polysaccharide mixtures separate into two phases when the salt concentration exceeds 0.1 M. Proteins and carboxyl- containing polysaccharides have a limited compatibility when either the pH exceeds the protein’s IEP (at any ionic strength) or the pH is equal to or less than the protein’s IEP and the ionic strength exceeds 0.2 M. With sulfated polysaccharides, globular proteins are usually incompatible at ionic strengths above 0.3 M, irrespective of the pH (Antonov et al., 1975, 1979,


1980; Grinberg and Tolstoguzov, 1972,1997; Tolstoguzov et al., 1985; Tolsto- guzov, 1978, 1988b, 1997a,b).

6. Effect of Molecular Size. Excluded Volume Effects

When attraction between nonidentical macromolecules (e.g., between proteins and polysaccharides) is inhibited, the major factor determining phase separation threshold is the excluded volume that depends upon the size and shape of the macromolecules (Semenova et al., l990,1991a,b; Tolstogu- zov, 1978, 1991, 1992).

Because molecules are not penetrable by each other, the volume of a solution occupied by a macromolecule is not accessible to other macromolecules. Therefore, a minimal distance between two adjacent spherical molecules of a globular protein equals the sum of their radii, or the diameter of one of them. This signifies that around each protein molecule there is an excluded volume not accessible to other molecules. This excluded volume is more than eightfold greater than the protein molecule itself since it includes the hydration water attached to this protein molecule. It is significantly greater for nonspherical macromolecules and depends upon the flexibility of the macromolecular chain, and its configurational, rotational, and vibra- tional properties (Tanford, 1961). The effects of spatial limitations are enhanced by the transition from a dilute mixed solution, where individual macromolecules are independent of each other, to a semi-dilute biopolymer solution where molecules interact and compete for the same space. Phase separation usually takes place when the total concentration of a mixed biopolymer solution exceeds a certain critical value corresponding to the regime of semi-dilute solutions. The phase behavior of biopolymer mixtures can be predicted from the excluded volume of the macromolecules (Semen- ova et al., 1990,1991a,b; Tolstoguzov, 1991,1992). Phase separation occurs at about 1-3% for mixtures of rigid, rod-like polysaccharides; about 2-4% for mixtures of linear polysaccharides with proteins of unfolded structure, such as gelatin or casein; about 4% or higher for globular protein- polysaccharide mixtures; and exceeds 10% for mixtures of globular proteins (Tolstoguzov, 1978, 1988c, 1990, 1991).

1. Protein Mixtures

From the viewpoint of structure, size, and molecular weight, globular proteins are macromolecular compounds, but are not typical polymers. Nor- mally, mixed solutions of classical polymers tend to be completely separated into nearly pure phases containing individual polymers. This reflects a very low co-solubility of classical flexible chain polymers. Unlike mixtures of


flexible chain polymers in a common solvent, significantly (at least ten- fold) higher phase separation threshold values, a high phase diagram asymmetry and quite similar phase behavior of their mixtures are typical of mixed protein solutions. Table I shows critical point coordinates for several protein pairs. Figure 3 gives phase diagrams for mixed solutions of some proteins of different conformations. The remarkably higher co-solubility of proteins compared to that of protein-polysaccharide mixtures seems to be of importance for biological functions of proteins, especially enzymes and enzyme inhibitors. The compactness, rigidity, rounded shape, and limited number of accessible ionizable side groups of a protein molecule make the phase behavior of proteins greatly different from common polymers and polyelectrolytes. Besides a relatively low excluded volume effect and polyelectrolyte nature, a significant difference in size (molecular weight) is probably of importance to higher co-solubility of globular proteins, especially those belonging to the same class within the Osborne classification. For instance, the small size of molecules of low molecular weight proteins,

C r

B

5 10 15

Gelatin, YO

D

Ovalbumin, YO Casein, YO

FIG. 3 Phase diagrams of some protein mixtures. A-Gelatin + legumin (pH 7.0, 40°C); B- Gelatin + legumio (pH 7.0, 0.5 M NaCl, 40°C); C-Ovalbumin + casein (pH 6.6, 20°C); D- Soybean globulins + casein (pH 6.9,25”C). .-critical point.


such as trypsin inhibitor, seems to be responsible for an increase in their compatibility with other proteins. This is probably due to an increase in mixing entropy of particles with similar shape and surface but differing in size. One more specific feature, which could contribute to high phase separation thresholds of globular proteins, is that their chemical information is mainly concealed in the hydrophobic interior of the protein globule. The tertiary structure of coiled and pleated polypeptides mimics on a molecular level that of proteins by hiding their chemical differences inside the globules. This molecular (or thermodynamic) mimicry minimizes chemical differences between proteins with respect to their interactions with both other macromolecules and solvent water. This results in better solubility and especially co-solubility of proteins in aqueous media. The same mechanism of thermodynamic mimicry is also probably used by polysaccharides forming helical structures and by polynucleotides coated by proteins to form viruses. The polyelectrolyte nature of biopolymers enhances co-solubility due to the contribution of low molecular weight counterions to an increase in mixing entropy. However, the effects of pH, salt concentration, and temperature on compatibility of proteins are less pronounced than for linear polyelectrolytes. Compatibility of proteins usually increases with ionic strength. Thermal denaturation of proteins and association or dissociation of oligomeric proteins greatly affects protein compatibility. Thermal denaturation can result in incompatibility of proteins of the same class according to the Osborne classification and even between the native and denatured forms of the same protein. The dissociation of protein complexes (such as casein arising from changes in pH, ionic strength, and binding of Ca2+ by complexons) and oligomeric proteins (e.g., seed storage globulins) results in an increase in compatibility. For instance, the compatibility of casein with soybean globulins and with pectin increases when the pH increases from 6.0 to 8.0 and sodium citrate is added (Anderson et al., 1985; Tolstoguzov, 19%; Polyakov er al., 1997).

2. Protein-Polysaccharide Mixtures

Generally, the chemical composition, structure, molecular weight and polydispersity of polysaccharides affect their co-solubility with each other and with proteins. Compatibility in protein-polysaccharide mixtures increases in the following order of polysaccharides: carboxyl- containing>neutral>sulfated; and in the following order of proteins: albumins>globulin>casein>prolamines>glutelins. These trends are observed in spite of marked differences in chemical composition, structure and molecular weight of the biopolymers. A marked similarity in phase behavior of protein mixtures with neutral and anionic polysaccharides reflects the formation of inter-biopolymer protein-neutral polysaccharide


complexes at low ionic strengths and pH differing from the protein’s IEP. Dissociation conditions correspond to those of phase separation. Generally, protein-polysaccharide mixtures show a lower critical phase separation temperature (Grinberg and Tolstoguzov, 1972, 1997; Tolstoguzov et al., 1985; Varfolomeeva et al., 1980).

An increase in excluded volume (molecular weight) of proteins and polysaccharides results in a decrease in compatibility. Biopolymer incompatibility increases under conditions favorable for enlargement of macromolecules, for instance, under a statistical coil-helix conformational transition. Linear polysaccharides (e.g., pectin and alginate) are less compatible with proteins than branched polysaccharides (e.g., gum arabic) of the same molecular weight. This obviously reflects the smaller excluded volume of branched macromolecules compared to linear molecules of the same molecular weight (Tolstoguzov, 1991).

Biopolymer incompatibility decreases with partial hydrolysis of macromolecules. Strong changes in compatibility resulting from a limited proteolysis (Danilenko et al., 1992) could be of interest for modeling the enzymatic activation of inactive forms of many proteins, such as trypsinogen (secreted form) modified by an enzyme to yield trypsin.

3. Phase Diagram Asymmetry

The competition between macromolecules for space in a mixed solution determines both the critical conditions of phase separation and the water and biopolymer partition between the system phases, i.e., phase diagram asymmetry. Normally, phase diagrams of biopolymer mixture are markedly asymmetric concerning compositions of the coexisting phases. Phase diagram asymmetry can be evaluated by the ratio of the critical point coordinates, by the angle made by the tie-lines with the concentration axis of one of the biopolymers and by the length of the segment of a binodal curve between the critical point and the phase separation threshold (Tolstoguzov, 1986, 1988b, 1991). The role of the size of macromolecules is quite evident from Table I and Fig. 3. The concentration of a smaller size biopolymer (protein) is higher at the critical point. The critical point is usually shifted from the phase separation threshold toward the axis of the biopolymer with lower hydrophilicity. The binodal is always closer to the concentration axis of the biopolymer of lower excluded volume. The greater the difference in the biopolymer molecular masses, the greater the shift of the binodal towards the concentration axis of the lower molecular mass biopolymer or the greater the shift of the phase separation threshold from the coordinate angle bisector in the same direction. Self-association of macromolecules can change both the excluded volume effects and the affinity of supermolecular


structural domains for the solvent water. As a result, the slope of tie-lines changes with an increase in the bulk biopolymer concentration.

4. Concentration and Fractionation of Biopolymers

The greater the phase diagram asymmetry the larger the difference in water content between the phases. The water content is higher in the phase of a more hydrophilic biopolymer with a larger excluded volume. Figure 2b shows that phase separation is accompanied by an increase in the concentration of each biopolymer in the coexisting phases. Protein concentration in phase D is higher than in the initial protein solution B and the mixed solution C1. The initial polysaccharide solution (Al) is diluted to point E. Phase separation can give rise to a strong increase in the concentration of one of the biopolymer phases and a corresponding dilution of the other phase. This phenomenon underlies a new method for concentrating biopolymer solutions called “membraneless osmosis.” Membraneless osmosis occurs between immiscible solutions. Therefore, the semipermeable mem- brane used for conventional osmosis is replaced by an interfacial surface between the immiscible solutions. Membraneless osmosis was first em- ployed to concentrate skimmed milk proteins using apple pectin. Figure 4a shows that a mixture of skimmed milk and 1% pectin solution breaks down into two liquid phases with the protein-rich phase containing about 20% of milk casein (Antonov et al., 1982; Tolstoguzov et al., 1985; Tolstogu- zov, 1988a,c, 1995,1996b, 199%; Zhuravskaya et al., 1986). Phase separation also enables biopolymers to be fractionated. Two approaches are used. The first is based on the phase separation threshold as determined by the less compatible of the biopolymer components. Partition of the other components corresponds to their relative affinities for the two phases and the interfacial layer. The second approach is based on the difference in the phase separation threshold of various proteins with the same polysaccharide. Accordingly, stepwise addition of a polysaccharide may induce separation and fractionation of proteins. For instance, the purification of baker’s yeast proteins from nucleic acids and some lipids to be used for human consumption was achieved by gradual addition of pectin (Bogracheva et al., 1983; Tolstoguzov, 1986).

5. Effect of Gelation

Phase equilibrium can only be achieved when liquid-liquid (Figs. 2,3, and 4a) phase separation occurs (Tolstoguzov, 1988b, 1995). Figure 4b shows the competition between phase separation and lyotropic gelation, i.e., from separation of the solvent. The equilibrium between the phases is not achiev-

PHASE DIAGRAMS FOR AQUEOUS SYSTEMS

1 Skimmed milk + polysacchandes High ester

Gum arabic. %wt B

1.6 1

0.6 20 40

10 20 30 Protein, %wt Protein Kwt.

21

Gelatin + Starch

Starch, %wt Starch, %wt

10

Gel- point

1

1 s 10 1s Gel- + lo Gelatin, wt polnt Gelatin, %wt

FIG. 4 Typical phase diagram for (A) an equilibrium protein-polysaccharide-water system (skimmed milk + high ester pectin) and systems with gelation of (B) the dispersed phase (skimmed milk + gum arabic), (C) the dispersion medium (gelatin + starch at 45°C) and (D) both phases (gelatin + starch at 35°C).

able when one or both phases of the system are gelled. Gelation stops the partitioning of the solvent and some solutes between the phases. Figures 4b, 4c and 4d illustrate phase diagram anomalies corresponding to gelation of the dispersed phase, the dispersion medium, and both phases, respectively. These anomalies can be observed, e.g., during phase separation of casein-polysaccharide (arabic gum and arabinogalactan) and gelatin-starch systems (Grinberg el al., 1971; Antonov er al., 1982). Gelation of the dispersed phase interrupts balancing of the osmotic pressure between coexisting phases. This is due to the back pressure of the gel stopping membraneless osmosis, water partition and establishing an osmotic equilibrium between the phases. The tie-lines converge to a certain point corresponding to the composition of the gelled dispersed phase. Gelation of both phases has the same consequences: an incomplete separation of the phases and nonequilib- rium with respect to the biopolymer distribution. The dispersed particles act as filler in the matrix (continuous) gelled phase.


IV. Features of the Cornpos-Won-Property Relationships in Mixed Biopolymer Systems

A. Low Interfacial Tension

The first feature of two-phase biopolymer systems is their very low interfacial tension. This results from both close compositions of the coexisting phases whose main component is the solvent, water, and a strong biopolymer co-solubility (Tolstoguzov ef al., 1974b; Tolstoguzov, 1994a). Low interfacial tension favors the stability of water-in-water emulsions.

B. Interfacial Layers in Biphasic Biopolymer Systems

The next feature of two-phase biopolymer systems is the presence of interfacial layers of low biopolymer concentration, low density, and viscosity, discussed as well in the chapter by Cabezas in this volume. The formation of interfacial layers results from unfavorable interactions of incompatible macromolecules. This reflects trends towards surroundings of the same type for each macromolecule. Nonidentical macromolecules are mutually depleted from the contact area between aqueous phases. The presence of an interfacial (or depletion) layer with a thickness of the order of magnitude of the size of the macromolecule and macromolecular aggregate has been demonstrated by both electron-microscopy of gels formed from a two- phase gelatin-dextran-water system (Tolstoguzov ef al., 1974a) and the measurement of steady-state viscosity in a two-phase liquid casein-sodium alginate-water system (Suchkov er al., 1981). Electron microscopy revealed an interesting interface-induced structure formation in gelatin gels filled with droplets of dextran-rich phase. Within the interfacial layer between the droplets of dextran phase and an isotropic network of the bulk of the gelatin gel, all fibrous aggregates forming the gel network are oriented normally to the surface of the droplets (Tolstoguzov ef al., 1974a). This structure-forming effect of dispersed particles filling the gel, probably reflects the minimization of contacts between immiscible biopolymers.

The interfacial (or depletion) layers are responsible for depletion flocculation and coalescence of dispersed droplets of water-in-water emulsion as well as for an interfacial adsorption of macromolecules and colloidal particles between the aqueous phases (sections IV C and D).

C. Probable Surfactant for Water-in-Water Emulsions

Compounds such as glycoconjugates and synthetic protein-polysaccharide conjugates could be adsorbed at the waterlwater interfaces and play the role


of surfactants for biphasic protein-polysaccharide systems. Macromolecules comprising two or several classes of incompatible biopolymers (e.g., polysaccharides or heterooligosaccharides covalently bound to the polypeptide chain) can have an affinity for both coexisting biopolymer phases. Adsorp- tion of protein-polysaccharide conjugates (as surfactants for water-water emulsions) can increase adhesion between the aqueous protein-rich and polysaccharide-rich phases, improve thermodynamic stability of water- water emulsions and the co-solubility (compatibility) of biopolymers (Tol- stoguzov, 1993a,c, 1994b).

The attraction and repulsion between the protein and polysaccharide parts of a hybrid macromolecule are demonstrated by its folding (collapse) and unfolding. This could greatly change the shape and size of protein- polysaccharide hybrids, their excluded volume, solubility, co-solubility, and the viscosity of their solutions and the interfacial layers. Attraction and repulsion between the protein and polysaccharide parts of hybrid macromolecules and control of their nonspecific inter- and intramolecular interactions could be used to model the folding-unfolding (denaturation- renaturation) of globular proteins and biological functions of glycoconjugates. Accordingly, surfactants for water-in-water emulsions in which both phases are protein-rich could be proteins with widely differing amino acid compositions or their aggregates.

D. interfacial Adsorption of Lipids

Lipid dispersed particles added to water-in-water emulsions may be either encapsulated by the protein-rich phase (section El ) andor concentrated at the interface between the two aqueous phases. Concentration and coalescence of lipid droplets at the interfacial layers can result in thin lipid layers between aqueous phases. Thus, interfacial adsorption of lipids in a water- in-water emulsion provides the physical basis for the formation of lipid protein two-dimensional layers and three-dimensional honeycomb-like lipid structures. Formation of a continuous lipid phase is used for large scale production of low-fat butter replacers based on gelatin-polysaccharide mixtures (Tolstoguzov, 1994a, 1996b).

E. Composition-Property Relationships in Single- and Two-Phase Mixed Biopolymer Solutions

Owing to competition for space occupancy and excluded volume effects, different macromolecules mutually affect each other’s behavior. This determines composition-property relationships in single-phase mixed solutions:

24 VLADIMIR TOLSTOGUZOV

the main reaction upon increasing biopolymer concentration in a mixed biopolymer solution is reduction of excluded volume effects. According to Le Ch2telier’s principle, an equilibrium system subjected to a perturbation reacts in a way that tends to nullify the effect of this perturbation. A reduction of excluded volume effects is achievable by a decrease in the size, concentration and mobility of biopolymer space-filling particles due to, e.g., compactization, self-association of macromolecules, formation of interbiopolymer complexes, gelation, and crystallization of biopolymers. An increase in excluded volume effects in biopolymer mixtures can have the following manifestations: (i) increase in thermodynamic activity, (ii) enhancement of association of biopolymers, (iii) reduction of co- solubility of macromolecules and phase separation, (iv) enhancement of phase diagram asymmetry, (v) enhancement of protein adsorption at the oiUwater and gaslwater interfaces, (vi) increase in the rate of gelation and reduction in the critical concentration for gelation of mixed solutions compared to solutions of the individual biopolymers, and (vii) increase in the crystallization rate of polysaccharides. Mixing of biopolymers can greatly (synergistically or antagonistically) change physicochemical properties of liquid and gelled systems. This relates to both bulk and surface properties of biopolymer solutions with the compositions lying on both sides of the binodal curve (Tolstoguzov and Braudo, 1983; Moms, 1990; Tolstoguzov, 1978, 1990, 1994a,b, 1995; Zayspkin et al., 1997).

1. Effect of Incompatibility on Protein Adsorption at the

The relationship between solubility and surface activity differs between biopolymers and surfactants of low molecular weight. For instance, metha- nol and acetic acid are mixable with water in any proportion. But when the length of the hydrocarbon radical is increased, solubility in water decreases and affinity for an oil phase increases. Such a chemical modification leads to surfactants (fatty acids and alcohols) of low molecular weight. By contrast, the solubility and surface activity of a biopolymer can be controlled without chemical modification by addition of an incompatible biopolymer. The least soluble biopolymer component of a mixed aqueous solution is the most rapidly adsorbed at the interfaces. The competitive adsorption of biopolymers could include a stage of phase separation in the microvolume at the oiYwater interface. Partition of a biopolymer between the bulk phase and the interfacial layer, including its adsorption at oil-water interfaces, can be managed by the addition of another biopolymer (Tolstoguzov, 1991). Additionally, complexing with lipids and/or anionic polysaccharides may reduce protein conformational stability, decrease solubility of the biopoly-

OWater Interface


mer (proteins and polysaccharides), increase biopolymer surface activity, and result in stronger emulsion stabilizing layers.

Adsorbed (on the oiYwater interface) and dissolved (in the dispersion medium) molecules of the same biopolymer may differ in their conformational arrangement and charge distribution. It was, therefore, assumed that adsorbed and dissolved molecules of the same protein cannot recognize each other as being the same and can exibit incompatible behavior with each other. Accordingly, the monolayer of already adsorbed protein molecules can inhibit further adsorption of dissolved molecules of the same protein. This would cause an extended characteristic plateau covering a wide concentration region between critical concentrations for monolayer and multilayer adsorption in the adsorption isotherm of a globular protein. This is not, however, the case in casein, i.e., a protein complex (of thermodynamically compatible subunits), where there is a continuous increase in the amount of adsorbed protein with an increase in concentration of the casein solution (Tolstoguzov, 1991, 1992, 1994a).

Addition of a polysaccharide can affect the stability of oil-in-water emulsions stabilized by protein in various ways (Burova et ai., 1992; Pavlovskaya et al., 1993; Tolstoguzov, 1991, 1992; Tsapkina et aL, 1992). An increase in protein adsorption on the oUwater interface can occur. This effect has been shown in mixtures of 11s broad bean globulins with dextran. Second is a decrease in the critical concentration required to produce multilayer protein adsorption with increasing amount and molecular weight of the polysaccharide. This results from phase separation of the continuous aqueous phase of the oil-in-water emulsion leading to a dispersed phase rich in protein-encapsulated lipid droplets. It has been shown that the stability of oil-in-water emulsions, stabilized by 11s broad bean globulin, is increased by addition of a polysaccharide that is incompatible with the protein. Third, depletion flocculation, i.e., reversible association of colloidal biopolymer particles within a nonwettable medium, i.e., a solution of incompatible biopolymer. Since colloidal dispersions are thermodynamically unstable, separation of mixed colloidal dispersions usually occurs at significantly lower concentrations than that of molecularly dispersed biopolymers (Tol- stoguzov, 1988c, 1997a).

2. Effect of Incompatibility on Confonnational Changes and

Since thermal denaturation of a globular protein is not accompanied by any substantial change in its molecular volume, addition of a polysaccharide does not markedly affect the conformational stability of proteins, but can greatly accelerate both aggregation of denatured molecules and gelation (Tolstoguzov, 1991; Burova et al., 1992). In mixtures, biopolymers behave

Gelation of Biopolymers


as if they were in a solution of a higher concentration. For instance, gelation of a 5% gelatin solution is greatly accelerated (from 7 days to 2 hours) and the elastic modulus of the gelatin gel increases by 2- to 5-fold due to addition of 0.1-0.2% of dextran or another polysaccharide, such as methylcellulose or agarose (Tolstoguzov et af., 1974a; Tolstoguzov, 1988a, 1990,1995; Zasyp- kin et al., 1997). Another example is amylopectin crystallization in gelatin gels. Doi (1965) showed that an increase in gelatin concentration from 16 to 46% reduces the time for complete crystallization of 0.5% amylopectin from 3 days to 20 hours at 4°C. Mechanical properties (e.g., the shear modulus and tensile strength) of filled gels are proportional to the volume fraction of the dispersed phase, i.e., they obey additivity laws (Gotlib et al., 1988; Suchkov et ai., 1988; Tolstoguzov, 1978, 1988b).

Biopolymers can be more “co-soluble” in the bulk of the gel than in the liquid phase. Normally, biopolymer gelation results in a decrease in excluded volume effects due to a decrease in the amount and mobility of space-filling particles. This makes the dispersion medium of the gel a better solvent than the initial mixed solution. Therefore, the gel dispersed phase in Fig. 4a can be regarded as a compartment with a better solvent quality than the bulk of the continuous liquid phase. This decrease in the excluded volume of biopolymers in the gel state results in the formation of interpene- trated networks by biopolymers immiscible in the liquid solution (Tolstogu- zov, 1990,1995). Presumably, for this reason, the rate of enzymatic hydrolysis of a solid (gel-like) food usually seems to be proportional to its volume rather than to its surface area.

V. Concluding Remarks

Phase behavior of biopolymer mixed solutions has been considered to illuminate the possibility that phase separation could occur in cytoplasm. There is already a large variety of experimental evidence for the limited thermodynamic compatibility of biopolymers and for phase separation under conditions typical of cytoplasm. The concentration of biopolymers in the cytoplasm of cells and extracellular biological systems is probably com- parable, or exceeds typical values of phase separation thresholds of biopolymers. Accordingly, two main aspects of incompatibility in biological systems are: (i) control of the thermodynamic activity of biopolymers and (ii) control of cytoplasm heterogeneity. Knowledge about composition-property relationships of aqueous biphasic systems may be of importance for a better understanding of the formation of cytoplasmic structures.

PHASE DIAGRAMS FOR AQUEOUS SYSTEMS

References

27

Albertsson, P.-A. (1958). Partition of proteins in liquid polymer-polymer two-phase systems. Nature (London) 177,709-71 1.

Albertsson, P.-A. (1972). “Partition of Cell Particles and Macromolecules.” Wiley- Interscience, New York.

kbertsson, P.-A. (1995). Aqueous polymer phase systems: Properties and applications in bioseparation. In “Biopolymer Mixtures” (S. E. Harding, S. E. Hill, and J. R. Mitchell, eds.), pp. 1-12. Nottingham University Press, Nottingham.

Anderson, O., Schmandke, H., Polyakov, V. I., Grinberg, V. Ya., Bikbov, T. M., Danilenko, A. N., Leontjev, A. L., and Tolstoguzov, V. B. (1985). Compatibility of gelatin with broad bean globulin in an aqueous medium. J . Food Sci. 50,1133-1135.

Antonov, Yu. A., Grinberg, V. Ya., and Tolstoguzov, V. B. (1975). Phasengleichgewichte in WasserEiweissPolysaccharid Systemen. Staerke 27,424-431.

Antonov, Yu. A., Grinberg, V. Ya., and Tolstoguzov, V. B. (1979). Thermodynamische As- pekte der Vertraeglichkeit von Eiweissen und Polysacchariden in waessrigen Medien. Nah- rung 23,207-214,597-610.

Antonov, Yu., Grinberg, V., Zhuravskaya, N., and Tolstoguzov, V. (1980). Liquid two-phase water-protein-polysaccharide systems and their processing into texturated protein products. J . Text. Stud. 11, 199-215.

Antonov, Yu. A., Grinberg, V. Ya., Zhuravskaya, N. A., and Tolstoguzov, V. B. (1982). Concentration of protein skimmed milk by the method of membraneless isobaric osmosis. Carbohydr. Polym. 2,81-90.

Antonov, Yu. A., Pletenko, M. G., and Tolstoguzov, V. B. (1987). Thermodynamic compatibility of polysaccharides in aqueous media. Vysokomol. Soedin., Ser. A 29,24774481,2482- 2486.

Asherie, N., Lomakin, A. and Benedek, G. (1996). Phase diagram of colloidal solutions. Phys. Rev. Lett. 77,4832-4835.

Beijerinck, M. W. (1896). Ueber eine eigentumlichkeit der loslichen starke. Centralb. Bakteriot. Parasitenkd. Infektionskr. 2, 697-699.

Beijerinck, M. W. (1910). Ueber Emulsionsbildung bei der Vermischung wasseriger losungen gewisser gelatinierender kolloide. Kolloid-Z. 7, 16-20.

Bogracheva, T. Ya., Grinberg, V. Ya., and Tolstoguzov, V. B. (1983). Ueber die thermodynamische Vertraeglichkeit von Gummiarabicum und makromolecularen Komponenten der Globulinfraktion der Baeckerhefe. Nahrung 27,735-740.

Broide, M. L., Berland, C. R., Pande, J., Ogun, O., and Benedek, G. B. (1991). Binary-liquid phase separation of lens protein solution. Proc. Natl. Acad. Sci. U.S.A. 88, 5660-5664.

Bungenberg de Jong, H. G. (1936). “La coaceruation, les coacervates et leur importance en biologie” Hermann, Paris.

Burova, T. V., Grinberg, N. V., Grinberg, V. Ya., Leontiev, A. L., and Tolstoguzov, V. B. (1992). Effects of polysaccharides upon the functional properties of 11s globulin from broad beans. Carbohydr. Polym. 18, 101-108.

Clewlow, A. C., Rowe, A. J., and Tombs, M. P. (1995). Pectin-gelatin phase separation: The influence of polydispersity. In “Biopolymer Mixtures” ( S . E. Harding, S. E. Hill, and J. R. Mitchell, eds.), pp. 173-191. Nottingham University Press, Nottingham.

Danilenko, A. N., Vetrov, V. Yu., Dmitrochenko, A. P., Leontev, A. L., Braudo, E. E., and Tolstoguzov, V. B. (1992). Restricted proteolysis of legumin of broad beans: Effect on thermodynamic properties of aqueous solutions and interaction with ficoll. Nahrung 36, 105-11 1.

28 VLADIMIR TOLSTOGUZOV

Dobry, A., and Boyer-Kawenoki, F. (1948). Sur l’incompatibilitk des macromolkcules en solution aqueuse, Bull. SOC. Chim. Belg. 57,280-285.

Doi, K. (1965). Formation of amylopectin granules in gelatin gel as a model of starch precipita- tion in plant plastids. Biochim. Biophys. Acru 94, 557-565.

Doi, K., and Nikuni, Z. (1962). Formation of starch granules from an aqueous gelatinous solution. Stuerke 14, 461-465.

Dubois. M., GiUes, K. A., Hamilton, J. K., Rebers, P. A., and Smith, F. (1956). Colorimetric method for determination of sugars and related substances. Anal. Chem. 28,350-356.

Durrani, C. M., Prystupa, D. A., Donald, A. M., and Clark, A. H. (1993). Phase diagram of mixtures of polymers in aqueous solution using Fourier transform infrared stectroscopy. Macromolecules 26,981 -987.

Edmond. E., and Ogston, A. G. (1968). An approach to the study of phase separation in ternary aqueous systems. Biochem. J . 109,569-576.

Flory. P. J. (1953). “Principles of Polymer Chemistry.” Cornell University Press, Ithaca, NY. Gotlieb, A. M., Plashchina, I. G., Braudo, E. E., Titova, E. F., Belavtseva, E. M., and Tolstogu-

zov, V. B. (1988). Investigation of mixed agarose-gelatin gels. Nuhmng, 32,927-937. Grinberg, V. Ya., Schwenke, K. D., and Tolstoguzov, V. B. (1970). Phase state of the gelatin-

dextran-water system. Izv. Akud. Nuuk SSSR, Ser. Khim., pp. 1430-1432. Grinberg, V. Ya., Tolstoguzov, V. B., and Slonimskii, G. L. (1971). Thermodynamic compatibil-

ity of gelatin and certain poly-D-glucanes in aqueous medium. Kolloid. Zh. 33,666-669. Grinberg, V. Ya., and Tolstoguzov, V. B. (1972). Thermodynamic compatibility of gelatin

and some D-glucans in aqueous media. Carbohydr. Res. 25,313-320. Grinberg, V. Ya., and Tolstoguzov, V. B. (1997). Thermodynamic incompatibility of proteins

and polysaccharides in solutions. Food Hydrocolloids 11, 145-158. Ismond, M. A. H., Murray, E. D., and Arntfield, S. D. (1988). The role of noncovalent

forces in micelle formation by vicilin from Vicio fubu. 111. The effect of urea, guanidine hydrochloride and sucrose on protein interactions. Food Chem. 29, 189-198.

Ismond, M. A. H., Georgiou, C., Amtfield, S. D., and Murray, E. D. (1990). Role of noncovalent forces in micellization using legumin from Viciufubu as a study system. J. Food Sci. 55,1638- 1642.

Kalichevsky, M. T., and Ring, S. G. (1987). Incompatibility of amylose and amylopectin in aqueous solution. Carbohydr. Res. 162,323-328.

Kalichevsky, M. T., Orford, P. D., and Ring, S. G. (1986). The incompatibility of concentrated aqueous solutions of dextran and amylose and its effect on amylose gelation. Curbohydr. Polym. 6, 145-154.

Kasapis, S., Moms, E. R., Norton, I. T., and Gidley, M. J. (1993). Phase equilibria and gelation in gelatidmaltodextrin systems-2. Curbohydr. Polym 21, 249-259.

Khokhlov, A. R., and Nyrkova, I. A. (1992). Compatibility enhancement and microdomain structuring in weakly charged polyelectrolyte mixtures. Macromolecules 25, 1493-1502.

Krause, S. (1978). Polymer-polymer compatibility. In “Polymer Blends” (D. R. Paul and S. Newman, eds.), Vol. 1, pp. 16-113. Academic Press, New York.

Kwei, T. K., and Wang, T. T. (1978). Phase separation behavior of polymer-polymer mixtures. In “Polymer Blends’‘ (D. R. Paul and S. Newman, eds.), Vol. 1 , pp. 141-184. Academic Press, New York.

Laurent, T. C., and Ogston, A. G. (1%3). The interaction between polysaccharides and other macromolecules. The osmotic pressure of mixtures of serum albumin and hyaluronic acid. Biochem. J. 89,249-253.

Ledward, D. A. (1993). Creating textures from mixed biopolymer systems. Trends Food Sci. Technol. 4 402-405.

Liu. C., Lomakin, A., Thurston, G. M., Hayden, D., Pande, A., Pande, J.. Ogun, O., Asherie, N., and Benedek, G. B. (1995). Phase separation in multicomponent aqueous-protein solutions. 1. Phys. Chem. 99,454-461.


Lomakin, A., Asherie, N., and Benedek, G. B. (1996). Monte Carlo study of phase separation

Medin, A. S., and Janson, J.-C. (1993). Studies on aqueous polymer two-phase systems contain-

Morris, E. R. (1990). Mixed polymer gels. In “Food Gels” (P. Hams, ed.), pp. 291-359.

Ogston, A. G. (1937). Some observations on mixtures of serum albumin and globulin. Biochem.

Ostwald, W., and Hertel, R. H. (1929). Kolloidchemische reaktionen zwischen solen von eiweisskorpern und polymeren kohlehydraten, Kolloid-Z. 47, 258, 357.

Pavlovskaya, G. E., Semenova, M. G, Thzapkina, E. N., and Tolstoguzov, V. B. (1993). The influence of dextran on the interfacial pressure of adsorbing layers of 11s globulin Viciu fabu at the planar n-decanelaqueous solution interface. Food Hydrocolloids 7, 1-10.

Polyakov, V. I., Grinberg, V. Ya., Antonov, Yu. A., and Tolstoguzov, V. B. (1979). Limited thermodynamic compatibility of proteins in aqueous solutions. Polym. Bull. 1,593-597.

Polyakov, V. I., Grinberg, V. Y., and Tolstoguzov, V. B. (1980). Application of phase-volume- ratio method for determining the phase diagram of water-casein-soybean globulins system. Polym. Bull. 2,757-760.

Polyakov, V. I., Kireyeva, 0. K., Grinberg, V. Ya., andTolstoguzov, V. B. (1985a). Thermody- namic compatibility of proteins in aqueous media.-1. Nuhnlng 29, 153-160.

Polyakov, V. I., Popello, I. A,, Grinberg, V. Ya., and Tolstoguzov, V. B. (1985b). Thermody- namic compatibility of proteins in aqueous media.-2. Nuhrung 29, 323-333.

Polyakov, V. I., Grinberg, V. Ya., Popello, I. A., and Tolstoguzov, V. B. (1986a). Thermody- namic compatibility of proteins in aqueous media.3. Nuhrung 30,81-88.

Polyakov, V. I., Grinberg, V. Ya., Popello, I. A., and Tolstoguzov, V. B. (1986b). Thermody- namic compatibility of proteins in aqueous medium. Nuhrung 30,365-368.

Polyakov, V. I., Grinberg, V. Ya., and Tolstoguzov, V. B. (1997). Thermodynamic incompatibility of proteins. Food Hydrocolloids ll, 171-180.

Poole, S., West, S. I., and Walters, C. L. (1984). Protein-protein interactions: Their importance in the foaming of heterogeneous protein systems. J. Sci. Food Agric. 35,701-711.

Popello, I. A., Suchkov, V. V., Grinberg, V. Ya., and Tolstoguzov, V. B. (1990). LiquidAiquid phase equilibrium in globulin/saltlwater systems. Legumin. J. Sci. Food Agric. 51,345-353.

Popello, I. A., Suchkov, V. V., Grinberg, V. Ya., and Tolstoguzov, V. B. (1991). Liquidhiquid phase equilibrium in globulin/salt/water systems. Vicilin. J. Sci. Food Agric. 54, 239-244.

Popello, I. A., Suchkov, V. V., Grinberg, V. Ya., and Tolstoguzov, V. B. (1992). Effects of pH upon the liquid-liquid phase equilibria in solutions of legumins and vicilins from broad beans and peas. Food Hydrocolloids 6, 147-152.

Samant, S. K., Singhal, R. S., Kulkami, P. R., and Rege, D. V. (1993). Protein-polysaccharide interactions: A new approach in food formulations. Int. J. Food Sci. Technol. 28,547-562.

Semenova, M. G., Bolotina, V. S., Grinberg, V. Ya., and Tolstoguzov, V. B. (1990). Thermody- namic incompatibility of the 11s fraction of soybean globulin and pectinate in aqueous medium. Food Hydrocolloids 3, 447-456.

Semenova, M. G., Bolotina, V. S., Dmitrochenko, A. P., Leontiev, A. L., Polyakov, V. I., Braudo, E. E., and Tolstoguzov, V. B. (1991a). The factors affecting the compatibility of serum-albumin and pectinate in aqueous medium. Curbohydr. Polym. 15,367-385.

Semenova, M. G., Pavlovskaya, G. E., and Tolstoguzov, V. B. (1991b). Light scattering and thermodynamic phase behaviour of the system 11s globulin-k-carrageenan-water. Food Hydrocolloids 4,469-479.

Suchkov, V. V., Grinberg, V. Ya., and Tolstoguzov, V. B. (1981). Steady-state viscosity of the liquid two-phase disperse system water-casein-sodium alginate. Curbohydr. Polym. 1, 39-53.

in aqueous protein solutions. J . Chem. Phys. 104, 1646-1656.

ing agarose. Curbohydr. Polym. 22, 127-136.

Elsevier, London.

J. 31,1952-1957.


Suchkov, V. V.. Grinberg, V. Ya., Muschiolik, G., Schmandke, H., and Tolstoguzov, V. B. (1988). Mechanical and functional properties of anisotropic geleous fibres obtained from the two-phase system of water-casein-sodium alginate. Nahrung 32, 661-668.

Suchkov. V. V.. Popello, I. A., Grinberg. V. Ya., and Tolstoguzov, V. B. (1990). Isolation and purification of 7s globulin and 11s globulin from broad beans and peas. J . A@. Food Chem. 38,92-95.

Suchkov, V. V., Popello, I. A., Grinberg, V. Ya., and Tolstoguzov, V. B. (1997). Shear effects on phase behaviour of the legumin-salt-water system. Modelling protein recovery. Food Hydrocolloids 11, 135-144.

Tanford, C. (1961). “Physical Chemistry of Macromolecules,” pp. 192-202. Wiley, New York. Tolstoguzov. V. B. (1978). “Artificial Foodstuffs.” Nauka, Moscow. Tolstoguzov, V. B. (1986). Functional properties of protein-polysaccharides mixtures. I n

“Functional Properties of Food Macromolecules” (J. R. Mitchell and D. A. Ledward, eds.), pp. 385-415. Elsevier Appl. Sci., London.

Tolstoguzov, V. B. (1988a). Creation of fibrous structure by spinneretless spinning. In “Food Structure-Its Creation and Evaluation” ( J . M. V. Blanshard and J. R. Mitchell, eds.), pp. 181-1%. Butterworth, London.

Tolstoguzov, V. B. (1988b). Concentration and purifzation of proteins by means of two-phase systems. Food Hydrocolloids 2, 195-207.

Tolstoguzov, V. B. (19%). Some physico-chemical aspects of protein processing into foodstuffs. Food Hydrocolloids 2,339-370.

Tolstoguzov, V. B. (1990). Interactions of gelatin with polysaccharides. In “Gums and Stabi- lisers for the Food Industry” (G. 0. Phillips, P. A. Williams, and D. J. Wedlock, eds.), Vol. 5. pp. 157-175. IRL Press, Oxford.

Tolstoguzov, V. (1991). Functional properties of food proteins and role of protein- polysaccharide interaction. Food Hydrocolloids 4,429-468.

Tolstoguzov, V. (1992). The functional properties of food proteins. In “Gums and Stabilisers for the Food Industry” (G. 0. Phillips, P. A. Williams, and D. J. Wedlock, eds.), Vol. 6, pp. 241-266. IRL Press, Oxford.

Tolstoguzov, V. (1993a). Functional properties of food proteins. Role of interactions in protein systems. In “Food Proteins. Structure and Functionality” (K. D. Schwenke and R. Mothes, eds.), pp. 203-209. VCH, Weinheh.

Tolstoguzov, V. (1993b). Thermoplastic extrusion-the mechanism of the formation of extru- date structure and properties. J . Am. Oil Chem. Soc. 70,417-424.

Tolstoguzov, V. ( 1 993c). Thermodynamic incompatibility of food macromolecules. In “Food Colloids and Polymers: Stability and Mechanical Properties” (E. Dickinson and P. Walstra, eds.), Spec. Publ. No. 113, pp. 94-102. Royal Society of Chemistry, Cambrige, UK.

Tolstoguzov, V. (1994a). Thermodynamic aspects of food protein functionality. In “Food Hydrocolloids Structure, Properties and Functions” (K. Nashinari and E. Doi, eds.), pp. 327-340. Plenum. New York.

Tolstoguzov, V. (1994b). Some physico-chemical aspects of protein processing into foods. In “Gums and Stabilisers for the Food Industry” (G. 0. Phillips, P. A. Williams, and D. J. Wedlock. eds.), Vol. 7, pp. 115-124. IRL Press, Oxford.

Tolstoguzov, V. (1995). Some physico-chemical aspects of protein processing in foods. Multi- component gels. Food Hydrocolloids 9,317-332.

Tolstoguzov, V. B. (1996a). Structure-property relationships in foods. In “Macromolecular Interactions in Food Technology” (N. Pams, A. Kato, L. K. Creamer, and J. Pearce, eds.). pp. 2-14. Am. Chem. SOC. Washington, DC.

Tolstoguzov, V. (1996b). Applications of phase separated biopoiymer systems. In “Gums and Stabilisers for the Food Industry” (G. 0. Phillips, P. A. Williams, and D. J. Wedlock, eds.), Vol. 8, pp. 151-160. IRL Press, Oxford.


Tolstoguzov, V. B. (1997a). Protein-polysaccharide Interactions. In “Food Proteins and Their Applications in Foods” (S. Damodaran and A. Paraf, eds.), pp. 171-198. Dekker, New York.

Tolstoguzov, V. (199%). Thermodynamic aspects of dough formation and functionality. Food Hydrocolloids 11,181-193.

Tolstoguzov, V. B., and Braudo, E. E. (1983). Fabricated foodstuffs as multicomponent gels. J. Tex. Stud. 14, 183-212.

Tolstoguzov, V. B., Grinberg, V. Ya., and Fedotova, L. I. (1969). Thermodynamic compatibility of gelatin and human serum albumin with some poly-D-glucans in aqueous medium. Izv. Akad. Nauk SSSR, Ser. Khim., pp. 2839-2841.

Tolstoguzov, V. B., Belkina, V. P., Gulov, V. Ya., Titova, E. F., Belavtseva, E. M., and Grinberg, V. Ya. (1974a). Phasen Zustand, Struktur und mechanische Eigenschaften des gelartigen Systems Wasser-Gelatin-Dextran. Staerke 26, 130-137.

Tolstoguzov, V. B., Mzel’sky, A. I., and Gulov, V. Ya. (1974b). Deformation of emulsion droplets in flow. Colloid Polym. Sci. 252, 124-132.

Tolstoguzov, V. B., Grinberg, V. Ya., and Gurov, A. N. (1985). Some physicochemical approaches to the problem of protein texturization. J . Agric. Food Chem. 33, 151-159.

Tombs, M. P. (1970a). Alterations to proteins duringprocessing and the formation of structures. I n “Protein as Human Food” (R. A. Lawrie, ed.), pp. 126-138. Butterworth, London.

Tombs, M. P. (1970b). Protein products. Ger. Pat. 1,940,561. Tombs, M. P. (1972). Protein products. Br. Pat. 1,265,661. Tombs, M. P. (1975). Aqueous protein composition. US. Pat. 3,870,801. Tombs, M. P. (1985). Phase separation in protein-water systems and the formation of

structure. In “Properties of Water in Foods” (D. Simatos and J. L. Multon, eds.), pp. 25-36. Martinus Nijhoff Publ., Dordrecht, The Netherlands.

Tombs, M. P., Newsom, B. G., and Wilding, P. (1974). Protein solubility: Phase separation in arachin-salt-water systems. Int. J. Pept. Protein Res. 6,253-277.

Tsapkina, E. N., Semenova, M. G., Pavlovskaya, G. E., Leontiev, A. L., and Tolstoguzov, V. B. (1992). The influence of incompatibility on the formation of adsorbing layers and dispersion of n-decane emulsion droplets in aqueous solution containing a mixture of 11s globulin from Vicia faba and dextran. Food Hydrocolloids 6,237-251.

Varfolomeeva, E. P., Grinberg, V. Ya., and Tolstoguzov, V. B. (1980). On the possibility of estimating weak interactions of macromolecules in solution from the experimental viscous flow activation energies data. Polym. Bull. 2, 613-618.

Walter, H., and Brooks, D. (1995). Phase separation in cytoplasm, due to macromolecular crowding, is the basis for microcompartmentation. FEES Lett. 361, 135-139.

Walter, H., Brooks, D. E., and Fisher, D. (1985). “Partitioning in Aqueous Two-phase Systems: Theory, Methods, Uses and Applications to Biotechnology.” Academic Press, Orlando, FL.

Walter, H., Johansson, G., and Brooks, D. E. (1991). Partitioning in aqueous two-phase systems: Recent results. Review. Anal. Biochem. l97,1-18.

Zasypkin, D. V., Braudo, E. E., and Tolstoguzov, V. B. (1997). Multicomponent biopolymer gels. Food Hydrocolloids 11,159-170.

Zhuravskaya, N. A., Kiknadze, E. V., Antonov, Y. A., and Tolstoguzov, V. B. (1986). Concen- tration of proteins as a result of the phase-separation of water-protein-polysaccharide systems, Nahrung 30,591-599,601-613.

compositions and phase diagrams for aqueous systems …...compositions and phase diagrams for...

Documents